Persia: An Open, Hybrid System Scaling Deep Learning-based Recommenders up to 100 Trillion Parameters

Author:

Lian Xiangru1,Yuan Binhang2,Zhu Xuefeng3,Wang Yulong3,He Yongjun2,Wu Honghuan3,Sun Lei3,Lyu Haodong3,Liu Chengjun3,Dong Xing3,Liao Yiqiao3,Luo Mingnan3,Zhang Congfei3,Xie Jingru3,Li Haonan3,Chen Lei3,Huang Renjie3,Lin Jianying3,Shu Chengchun3,Qiu Xuezhong3,Liu Zhishan4,Kong Dongying3,Yuan Lei3,Yu Hai3,Yang Sen3,Zhang Ce2,Liu Ji5

Affiliation:

1. Kwai Inc, Seattle , WA, USA

2. ETH, Zurich, Switzerland

3. Kuaishou, Beijing, China

4. Kuaishou, Beijng, China

5. Kwai Inc, Seattle, WA, USA

Publisher

ACM

Reference35 articles.

1. (n.d.) Avazu CTR. www.kaggle.com/c/avazu-ctr-prediction. (n.d.) Avazu CTR. www.kaggle.com/c/avazu-ctr-prediction.

2. (n.d.) Criteo CTR. ailab.criteo.com/download-criteo-1tb-click-logs-dataset. (n.d.) Criteo CTR. ailab.criteo.com/download-criteo-1tb-click-logs-dataset.

3. (n.d.) Parallel Distributed Deep Learning: Machine Learning Framework from Industrial Practice. https://www.paddlepaddle.org.cn/. (n.d.) Parallel Distributed Deep Learning: Machine Learning Framework from Industrial Practice. https://www.paddlepaddle.org.cn/.

4. (n.d.) Taobao CTR. www.kaggle.com/pavansanagapati/ad-displayclick-dataon-taobaocom. (n.d.) Taobao CTR. www.kaggle.com/pavansanagapati/ad-displayclick-dataon-taobaocom.

5. Jianmin Chen , Xinghao Pan , Rajat Monga , Samy Bengio , and Rafal Jozefowicz . 2016. Revisiting distributed synchronous SGD. arXiv preprint arXiv:1604.00981 ( 2016 ). Jianmin Chen, Xinghao Pan, Rajat Monga, Samy Bengio, and Rafal Jozefowicz. 2016. Revisiting distributed synchronous SGD. arXiv preprint arXiv:1604.00981 (2016).

Cited by 14 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Energy-Efficient Federated Knowledge Distillation Learning in Internet of Drones;2024 IEEE International Conference on Communications Workshops (ICC Workshops);2024-06-09

2. AtRec: Accelerating Recommendation Model Training on CPUs;IEEE Transactions on Parallel and Distributed Systems;2024-06

3. CAFE: Towards Compact, Adaptive, and Fast Embedding for Large-scale Recommendation Models;Proceedings of the ACM on Management of Data;2024-03-12

4. POSTER: Pattern-Aware Sparse Communication for Scalable Recommendation Model Training;Proceedings of the 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming;2024-02-20

5. Enhancing the Accuracy of 6T SRAM-Based In-Memory Architecture via Maximum Likelihood Detection;IEEE Transactions on Signal Processing;2024

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3